Fog Creek Software
Discussion Board

Why Napster failed to deliver

I can tell you exactly why Napster failed to deliver.  I interviewed there last year, and got an eyeful of the place.

They were in serious need of adult supervision.  Their criteria for hiring was essentially having a graduate level knowledge of CS, specifically complex algorithms.  They were more interested in your ability to decode fully obfuscated code than in generating code that was clear.

They had a roomful of young, bright code jockeys with academic backgrounds, when what they needed was a group of experienced get-it-done engineers who actually understood things like "if it's not done before the deadline, it doesn't matter how beautiful the architecture is".

Writing something as boring as a billing system is not something you apply algorithms gurus to, but the gurus already there were completely blind to that fact, and the lawyers actually running the company weren't paying attention.

Basically, what they needed was one competent manager in Engineering.

Glad I wasn't offered a job
Thursday, September 5, 2002

So what you're saying is that Napster doesn't exist anymore because they didn't hire people like you, and it had nothing to do with their legal troubles with the recording industry.


Luke Duff
Thursday, September 5, 2002

I don't think he said that's why they failed as a company, I think he said that's why they failed to deliver code.

I tend to agree -- every time I hear of one of these companies with a 19 year old CTO (Marc Andreessen, Shawn Fanning) I know they are in trouble. For some reason Silicon Valley companies think it's perfectly reasonable to make the young wiz kid who wrote the prototype into the CTO, and it always fails. I suppose it's politics; you can't tell the *inventor* that he's not going to be Chief Technical Officer.

That said, I'm sure Fanning is a great hacker and I'd hire him in a minute. But not as CTO.

Joel Spolsky
Thursday, September 5, 2002

No.  Napster failed to deliver a billing system on time for their paying customer Bertelsmann (who later attempted, and failed, to buy them) because they failed to hire people like me. :-)

They failed as a company because they failed to come up with a business model that generated revenue, like a great many other "if we just get big, we'll charge later" dot coms tried, and failed, to do.  Their model: antagonize the music industry, bully them into realizing there's money to be made on the Net, and get bought.  They succeeded in the first step, but not the second or third.

Glad I wasn't offered a job
Thursday, September 5, 2002

"For some reason Silicon Valley companies think it's perfectly reasonable to make the young wiz kid who wrote the prototype into the CTO, and it always fails."

Works better in Seattle though ; )

Robin Debreuil
Friday, September 6, 2002

Hmm. I though the reason they failed was that they /did/ deliver, and that was what attracted the attentions of the record companies.

Friday, September 6, 2002

I think regardless of who the programmers were, how many programmers they had, whatever development philosophies they followed, etc, there would still be no Napster today.

There was no engineering solution to Napster's problems.

Luke Duff
Friday, September 6, 2002

The key sentence in the original post:

"They were in serious need of adult supervision."

Napster certainly needed some on the legal front.  "It's OK to help others steal from companies no one likes" is a kid's idea of a business plan.  And, as the original poster suggests, maybe they needed adult supervision on the engineering side, too. 

Hardware Guy
Friday, September 6, 2002

Napster delivered Version 1 of their code (the file swapping system), and that caused all the ruckus that ultimately led to their demise.

However, Joel's original question on the front page was: Why didn't Napster deliver VERSION 2 (i.e., their for pay service) after working on it for two years.  The answer was given by the original poster. 

Napster didn't die because of their legal problems.  Napster died because it ran out of money.  It ran out of money, yes, because it had to pay all of those lawyers, but also because it had no revenue stream.  It had no revenue stream because they failed to deliver the for pay service, which might have generated enough money to keep paying their lawyers.  With a successful relationship with Bertelsmann, they might have fended off the rest of the music industry, and might have a contract house for them on the Net, generating a online for-pay music system.

James Montebello
Friday, September 6, 2002

I don't think Napster would have succeeded even if they did deliver a for-pay service with minimally intrusive (hah!) DRM. The record companies have a vested interest in supressing all non-traditional distribution channels. They would have refused to license most popular songs to Napster's service, eventually starving it to death. Any interest they showed in Napter's true potential was feigned.

This happened once before - see,1640,14814,FF.html

Dan Maas
Saturday, September 7, 2002

"I can tell you exactly why Napster failed to deliver."

So can I. It sucked. It didn't do what it's users (based on a sample of me and my friends) wanted and it didn't do what big business wanted.

Robert Moir
Saturday, September 7, 2002

I don't think it was a problem of delivering Version 2.0. I still haven't heared of any company making money with downloadable songs. Another large record company (I believe it was Universal) here in Germany just started their version of a pay-per-download service. It's around 1$/song, and they still include copyright mechanisms with that, so you can't do much with it (especially not copy it to CD so you can listen to it in your car). They say it is successful, but refuse to give _any_ numbers, so I guess it is rather not.
As long as there is not a single company around that can't even break-even with offering downloadable songs, there is no reason to believe that it was just the missing release of V2 that broke Napsters neck.

Monday, September 9, 2002


Version 1:  You get all kinds of music for free.  No copyright protection.

Version 2:  Same as 1.0 except now you have to pay for it.  And you can't copy it.

Excuse me?  Why would anyone "upgrade" to Version 2?  Even if they delivered it it would've failed.  Plus it wouldn't even be the same application.  Napster was a P2P with a few servers to organize things.  Version 2 would've required people to open up and waste their bandwidth distributing software for someone else?  It doesn't make sense... It was doomed to failure....

Now here's what I would like.  A service that allows me to download a song for cheap (i.e. 50 cents), has a huge library (basically 99.99% of the stuff out there), and allows me to keep d/ling the songs I already paid for "forever", in different formats and quality settings so I don't need to keep 30 Gigs of MP3's I never hear in my HD.  And no copy protections whatsoever, since it's useless anyway.

Do not use HTML tags
Monday, September 9, 2002

"I still haven't heared of any company making money with downloadable songs"

I paid to go to a concert on the back of songs d/l'd for free from the group's website A slightly different model maybe, but they still made money out of the deal.

Tuesday, September 10, 2002

I recall reading an interview with Courtney Love where she talked about alternatives for artists via online distribution.  Alot of it were ideas like using the songs as a venue to gain interest for concerts, etc.

Problem is, these venues often result in the removal of the middle man, IE the record industry, and thus the copyright holders, IE the record industry, is less then inclined to allow let alone promote these things.

But back to not delivering 2.0.  This sounds a lot like the same mistakes Netscape made with the 4.0 to the non-existant 5.0.  A Code-monkey way of developing with no concern for actual costs and engineering fundamentals.

Lucas Goodwin
Friday, September 13, 2002

*  Recent Topics

*  Fog Creek Home